Its pure guessology. Then again its no better or worse than any other model. The time depths involved in the seperation of these branches is huge and all we see are the tail ends of where these branches went to. It doesnt tell us where they spend many millenia in between. The seperation time of the branches leading to the SNP defined clades are very big indeed and in many cases far older than the SNPs. So there is a huge amount of wandering time that is very hard to put into a prose story. The map is very much a guess IMO. The branches leading to M269 and M73 and V88 split so long ago with such fundemental changes in economy and climate happening since then that it seems almost impossible to reconstruct the locations of their common ancestors. The evidence of this is very old interclades but relatively young intraclades.
My main feeling about R1b in general is it lacks the sort of bushy nature around the time of the early farming period that I would have expected if it was within the early farming area. Also the lack of R1b and R1a in early Neolithic samples to date. It seems to me to have been peripherall to farming until later in the Neolithic at least. On that basis I find placing it in the middle east, Anatolia etc unlikely unless of course it had a very special niche that kept it peripheral for a long time.
Personally I think there is too much of an attempt to make R1a and R1b contrast when they actually seem to share many characteristics such as unimpressive branching until late in the Neolithic, a lack of Neolithic ancient DNA and a major take-off late in the Neolithic. I suspect they actually were in a similar area and we are reading way too much into modern distributions caused by movements after 3000BC. They clearly took different paths but that does not mean they had radically different origins. Anthony pretty well outlines a model which involves most of the R1a associated cultures and language groups as having orignated in a string of cultures that had a kind of reflux movement back east from the Carpathian area through the forest steppe and eventually into Asia. This is all post-2700BC. It fits rather neatly and I think he is probably right on this.
However, this does not tell us anything about what was going on in the actual European steppe before this in terms of R1a-R1b relations and distribution. In fact you could say that Anthony's model for the R1a-cultures-languages group relating to Slavic, Baltic and I-Iranian works so well that it strengthens the idea that R1b is probably also linked to an (apparently earlier split) of the western IE peoples and there is a real genetic link.
It suggests to me that R1b probably headed west at some point between 5000-3000BC with the bulk of the centum group. The problem with finding a solution at the later end of this scale is simply it requires models that are very contrasting in their complexity and lack of self evident natutr than the R1a centum cultural model that Anthony presents.
I think Anthony is correct in his model of the eastward spread of the Saetem group through the forest steppe and ultimately to Asia but why then is the apparently equally convincing R1b-western IE model so hazy. The basic contrast is that R1a and the saetem group is presented as a folk movement while the apparently R1b related centum group (beyond the east of Europe) has always been a house of cards model that doesnt feel like a folk movement at all. I wonder if the way of resolving this could be that R1b and indeed the centum branch actually existed pre-4000BC in at least the easternmost part of the farming world and the attempt to link it to steppe cultures of later times is actually wrong.
Perhaps it dispersed west from somewhere in SE Europe around the Carpathians in the period 5000-4000BC prior to the major steppe horizons. Huge cultures like TRB (c. 4000-3000BC) for example could easily have been in a position to transfer words for inventions like the wheel across chunks of Europe and the time gap (centuries rather than millenia) and geography gap between the spread of this culture and the invention of the wheel was not so great IMO that a borrowing like that would be obvious from lingusitics. Clearly the original Renfrew model of the very earliest farmers being IE does not seem to work but I am not so sure that the mid-late Neolithic cultures like TRB etc can be ruled out as IE speaking on the basis of vocab.
The simplest solution to me would be that the R1a eastward spread associated with the eastern IE languages is indeed a copper age one of c. 2700BC onwards but that the western IE languages split off and spread west before this and were not assocaited with steppe cultures. Trying to squeeze steppe solutions into western IE languages always looks tenious IMO. Of course this does all lead back to the question of what the ultimate common denomenator of both groups (Proto-IE) was if it was not steppes nomads. I still feel there is a strong possibility that PIE was related to non-steppe (but close to the western edge of the steppes) groups that had been brought into farming well before the Yamnaya horizon.
I personally think the whole Bug-Dniester/Cuc-Type transition has something to do with R1b being brought into the farming zone with the Bug-Dniester group being the element most likely to have had R1b. Essentially they could be Bug-Dniester people gentically but who were transformed into a farming culture and spread west as a farming culture (with steppe influences) when the climatic decline that ruined C-Tryp and drove steppe elements west happened. Of course I realise that this moves the PIE world slightly west to the western edge of the steppes but I have never found the Uralic contact evidence anywhere near as clinching on a more easterly PIE homeland than it is often presented as - there are far too many unknowns.
So returning to the map, I dont believe R1a and R1b were in hugely contrasting locations in say 4000BC. Perhaps R1b was among the Bug Dniester people who were absorbed into farming by the Cuc-Tryp group while R1a was east-adjacent. I dont buy the whole idea that R1b was in the early farming zone of Anatolia and the middle east until much before 4000BC. It is important to remember that there is no such thing as an archiac branch, just the modern tail end of different branches of equal ultimate age. Also, remember that the area north of the Black Sea has had incredibly population change in even the last few centuries and many different groups dominated this area in the last 3000 years, not to mention the late Slavic expansions etc. So, this crucial area is essentially of no use in using modern populations as proxys for 5000 years ago.