When trying to find a particular file with a very common name (i.e. few word string commonly used) I find that there will be a mass of search results around the file I am actually looking for.

This appears to restrict the number of results for the specific file in question (I know this because I have had to conduct more accurate searches in the past to obtain better results).

how about a search system whereby the user locates the specific file, which may be amongst many others, and can then select to conduct a search looking only for further copies of the exact same file on the network???

This would also help filter out crap in a search when you are already downloading and would improve the refresh rate for files with more commonly composed names (as well as reducing the instances of adult content blah blah blah).