I think the (implicit) charge that the editors of top journals are keeping good papers out for snobbish reasons is mostly unfounded. I don't deny that their tastes shape what gets published, of course. But in most circumstances, they are not subject matter experts, and the first step of evaluating any paper is that is not an obvious desk-reject (for poor writing, crankery, etc.) is to send it to a group of relevant experts for "quick opinions." Only if these opinions are sufficiently positive is the paper sent for a full referee report (again by an expert), and in most (but not all) cases the referee's suggestion is followed. So the picture of editors just arbitrarily killing papers they don't like is not accurate.
Further, it's not like no finite combinatorics/graph theory gets published in these journals. Just not a lot, because it's not sufficiently interesting/valuable to the broader community. (Annals of Math almost published Hales's proof of the Kepler conjecture, after all; eventually a proof appeared in another top journal.)
Also, re: connections, you can easily check the author affiliations for papers in these journals. There are plenty of people from universities that are not so well known. It's hardly an "old boys club."
There was a notorious case that a paper by Wehring solving the largest outstanding question in lattice theory was rejected by JAMS despite glowing referee reports (and being pretty short). So it's pretty much an example of editors arbitrarily killing a paper. (It appeared in Advances in Mathematics.)
You can see a letter to the editor of the Bulletin about it. Look at the affiliations of the people protesting the decision, and the affiliations of the editors defending it: http://www.ams.org/notices/200706/tx070600694p.pdf
I don't really see a problem with this decision, though I am sympathetic to Wehrung. As the editors note, there are a ton of great papers JAMS doesn't publish (due to severe page count constraints at the journal). Their reasoning is not "arbitrary"; it was spelled out quite clearly for the authors in the rejection notice they got. Many other papers in "trendy" subjects face the same fate.
What do you think they say? "You're not well-connected, so we're rejecting you"? How do you think political decisions get defended in academia? For insiders, you listen to the referees. For outsiders, you invoke some vague criterion of lack of fit or lack of interest. And people can come along and say "If this field is important, why doesn't it appear in JAMS"? It's a self-fulfilling prophecy.
You seem determined to believe that the rejection was for "political" reasons despite having no evidence for this claim. The quoted reasoning was not "lack of fit" or "lack of interest," it was lack of "interaction with other areas of mathematics." Going back to my first comment, there's a general sense in the community that this sort of interaction among mathematical subfields (or with physics) is prized in research, and it's one of a few criteria that form the sociologically dominant view of what constitutes "good mathematics." You might disagree with these criteria, but it's not hard to see why that paper might lose out to other excellent papers when judged by them. This doesn't look like politics (favoring "insiders") to me; it looks like the consistent application of a value judgment about what good research is.
Again, whether you think the criteria should be the way they are is a separate conversation.
Further, it's not like no finite combinatorics/graph theory gets published in these journals. Just not a lot, because it's not sufficiently interesting/valuable to the broader community. (Annals of Math almost published Hales's proof of the Kepler conjecture, after all; eventually a proof appeared in another top journal.)
Also, re: connections, you can easily check the author affiliations for papers in these journals. There are plenty of people from universities that are not so well known. It's hardly an "old boys club."