Although I know of no fossil evidence that would preclude plesiosaurs from living or feeding in or around the edges of the Mesozoic equivalent of a mangrove swamp, I think that a long, relatively inflexible neck would be a hindrance for navigating in a highly congested and very complicated environment...
Unfortunately most people continue to view plesiosaurs as they were portrayed by almost all of the early artists... with a long, snake-like neck... Charles Knight probably did one of the more extreme versions...http://www.oceansofkansas.com/images2/knight3.jpg
...and as late as 1943, Sam Welles, one of the leading plesiosaur experts at the time, published drawings (fig. 10) of the extremes of movement that he envisioned for elasmosaur necks:
Welles, S. P. 1943. Elasmosaurid plesiosaurs with a description of the new material from California and Colorado. University of California Memoirs 13:125-254. figs.1-37., pls.12-29.
Although we are still learning about how plesiosaurs swam... (imagine driving your car while sitting at the end of a boom that extended 15-20 feet in front of the front wheels - e.g. steering and propulsion - I'd love to see some one do a 'flight simulator' program on an elasmosaur). I think it is a pretty safe bet that elasmosaurs were relatively slow moving and had a fairly large "turning radius"... something much more suited for open water than the narrow confines of a tree root congested swamp.
Also, please note that due to "global warming" during the Late Cretaceous.. the Earth was pretty much a water world... with the surface about 85 percent covered by oceans (compared with 71% today)... Lots of shallow open water that would have be great habitat for a fish eater...
My two cents.