I don't know exactly the history of discussion about migration of I2a-Din here on forum, so I am sorry if I say something old.
The new Mathieson's paper does not bring more light to history of this subclade. Or maybe yes, because the only sample with L147.2 mutation is not from Balkans, but it is Motala12 (is this a new information?). If I understand it well, this is the first aDNA evidence for I2a-Din. For me, this completely changes the concept of how this haplogroup arised and migrated.
I am not a very big fan of physical anthropology, I think we have now more sophisticated methods for studying human populations. But I always asked about how is it possible that tall "Borrebies" are so frequent variant in Northern Europe and Balkans? What does connect these autosomally different places? Both of them have Borreby crania and both of them are tallest populations in Europe, but common ancestor of northern I1 and southern I2 is too old to connect these places and phenotypes. HGs in Eastern Europe were significantly taller than WHGs, for example in this abstract. This was probably due to sharing more alleles with ANEs, which is also the case of Latvian HGs, SHGs and new Iron Gates HGs that were predominantly I2a2 and R1b with some I2a1, but not I2a-Din.
So, can't be Motala12-like (SHG) population the answer for this phenotype in Balkans? I2a-Din men migrated (God knows why and how) to Dinaric Alps from the north carrying autosomal variants for tallness and specific cranial structure and were replaced by I1 men in the north.
I realize this is only a speculation with no evidence, but very interesting I must say.So, let's discuss.