There's some truth to what you say, but I think it's a little over-cynical. I know the biology/chemistry world so that's all I can speak on. In bio/chem the peer review system has flaws, but it's largely effective, IMHO. I've never seen a paper take longer than 6 months from submission to publication (I've published 22 papers) and the majority take ~3 months. There are bad reviewers out there that are unscrupulous, lazy, or incompetent... but there are also plenty of good ones and a good editor can tell them apart and adjudicate appropriately.
As to the purpose of peer review and whether it's still needed: Biology/chemistry is huge and there are probably 100-200 journals that each publish ~50 papers/month that are potentially relevant to my work. I skim them all (and use e-mail alerts to notify me of anything especially relevant) but I certainly don't have time to read them all with sufficient depth to be critical. As such, I rely on editors and reviewers to pre-screen papers and only let the highest quality research into the most-prestigious journals. The system is, of course, flawed in many ways... but, if nothing else, the competitive-ness of the top-journals "keeps the riff-raff out". With the exception of articles that are directly relevant to me, I only take a close look at papers that appear in the top ~5 journals because most of the lower-tier journals are pretty boring. My behavior is pretty typical, from what I've seen in my field.
In a perfect world I'd have enough time to evaluate everything on it's own merit, but I don't (no-one does). But if a paper has made it into a good-to-great journal, that at least means that (a) an editor thought it was interesting and (b) three or four respected scientists read it carefully and thought it was pretty good. As you pointed out, there are lots of places where this system can err, but overall it does a pretty good job and serves my needs well. It's sort of like the point-system on HN: good comments sometimes get missed, and bad comments sometimes get up-voted... but overall it's a pretty effective way to help people separate the interesting from the banal or trivial.