Well, I myself believe that the most likely origin of the Albanian language is in late IA Illyrians, possibly having also absorbed some Daco-Thracian people and got some influence from them, basically those who lived near what is now Kosovo (North Albania, Kosovo, South Serbia)... But I still don't think it's technically correct to say "they descend from Illyrians, because most of them have the same Y-DNA haplogroups, so case closed". R1b-Z2103, J2B-L283 and E1b-L618 all predate the actual Illyrian population by millennia, so they may even be associated with them, but they certainly do not depended on the Illyrians alone to exist and expand. It's totally possible that some other neighboring ethnicities, beside the Illyrians, also had the same lineages.
Besides, Y-DNA haplogroups can easily boom or bust in frequencies and are so subject to genetic drift that it's totally possible, perhaps even likely that even in the absence of any major admixture with other people (which is also unlikely over millennia) the Y-DNA distribution will become different from the ancient Y-DNA distribution after 2000 or 3000 years.
Additionally, of course there is also the fact that we just don't know if those aDNA samples were "Illyrian" (well, Illyrian in an ethnic/cultural sense they most certainly were not, maybe "pre-proto-Illyrian" or something). This hypothesis assumes as a fact that not just genetic, but also complete ethnocultural continuity happened in the western Balkans since basically the Late Neolithic, whereas these lineages might've been just assimilated by Illyrians.
So, I think these Y-DNA are an evidence, but certainly not an unquestionable proof as the OP seems to think.