Forgive me for spoiling the hopeful mood, but I'm not so sure the uneditable review idea would work.
The obvious reason is that in order to write a review about a wikipedia article, you'd have to be interested in wikipedia in the first place. So if I'm an academic interested in moving wikipedia along, why should I bother with a review? It would be less work to make the changes myself. My name would automatically be associated with the edit on the article's history page. And even if I write such a review, people would read it and fix the mistakes and missing issues I found (talking about duplicated effort here!). So, soon after my review is out, it won't fit the article anymore, because the article changed. So, all people would soon find is my outdated (=incorrect) review, with my name below it. No thanks!
As a to-be-academic, I'd rather have a stable article that says "...based on [[this article]] at wikipedia, edited by ..." (or "reviewed by" or "streamlined by";) where there's a backlink to the wikipedia article, maybe like "For a more current, but unreviewed version, see [[]]".
I might be wrong, and this is the "magic formula", but I don't see many academics interested in that review function, certainly less than in the original Nupedia (and even that didn't work...)
Magnus