Are you saying the "British Journal of Clinical Pharmacology" is not a peer-reviewed journal? But even then, Heuberger successfully defended his PhD.
I agree with most of your points, and over time have made these very same points.
When deciding which substance are banned, WADA only concerns itself with a "potential to enhance" performance. WADA has no interest to actually prove performance enhancement for any, let alone all Olympic sports. They just need enough to make a reasonable subjective one-sided, unchallengeable assessment.
And I agree a drug that "works" for archery, or cycling, or sprinting, may not necessarily "work" for distance running.
Heuberger's time-trial found that EPO cyclists "performed" in a lab as expected, but this did not translate to a real world time trial "performance" up the Mont Ventoux, contrary to expectations. The EPO cyclists were (insignificantly) slower, despite the improved lab parameters.
That seems like an important result. The conclusion here is that realities found in the lab do not necessarily describe the reality on the track or the roads. This is what most studies say in their limitations sections.
I'm not sure what I took out of context, but whether or not I take his research out of context, the proper way to rebut these points would be to come up with disproving counter examples. For example, when he concludes, after conducting a literature review, that studies look at the wrong things on the wrong people, if he is wrong then, we should be able to find one or more studies who looked at the right things on the right people. When he says that few studies use control groups, and/or are double-blinded -- this can be verified or contradicted, whether the statement is peer-reviewed in a journal, or not.