Indeed, at least with flies the organismal complexity is a bit higher. I just find the work fascinating. The real way to test this stuff is just to try the overexpressions in other models, and if they don't yield similar effects, move on. Rather than getting bogged down chasing a ghost.