The problem with this is it doesn't offer compelling evidence that twin studies are biased, it just assumes they are because the other studies are consistent with one another, just like the twin studies are with each other. It just flips the assumption without explaining anything.

The equal environments assumption has been tested again and again and it doesn't hold up either. Twins who are assumed to be MZs but are actually DZs and vice versa look like their biological relatedness, not their assumed relatedness. And factors involved in appearance (e.g., attractiveness) turn out not to affect heritability estimates either.