Broadly speaking, the omitted results are any URIs that have an identical URL to a result already given, but a different querystring.
There are other reasons a result would be in that list, but they are all based on properties of the resource, not the content.
Think about it. If two web sites list the same information on you, and google spidered them and determined that to be the case (which their system can't do, because it doesn't think), would they drop one site from their index and only list the other? No, because that would make for a crap search engine.