Heh, happy to take the blame/accolades for the inspiration! :) I definitely think there's room for improvement (no particular experience with PNAS) in general

Heh, happy to take the blame/accolades for the inspiration! :) I definitely think there's room for improvement (no particular experience with PNAS) in general -- peer review is very flawed, and there are huge problems with publication bias, the area being so specialized that the reviewers (indirectly) know whose article they're reviewing, etc. Actually, I think one of the most interesting things (science-wise) to come out in recent years is the meta-science of science-about-science, fuzzy as that may be. I'm guessing you already know of him, but Ben Goldacre has been a great source (for us outsiders) for information on this.