A few months ago I wrote an article for Macworld titled The Real Paperless Office [1], and it received tons of positive feedback. Basically I explained how to scan documents in such a way that you end up with a PDF containing both a high-res image of the original and searchable text (PDF+text), meaning that in many cases you can ditch the original documents. My preferred tool for the scanning portion of the job was a Fujitsu ScanSnap scanner; the one I brought with me to France was the original Mac model, an fi-5110EOXM.
Just after that article was published, Fujitsu introduced a new ScanSnap model, the S510M, which I recently reviewed for Macworld [2]. I'm delighted to say that it makes scanning to PDF+text even easier than before, and avoids the need some people had for workarounds like AppleScript Folder Actions that I'd described in my article. (I still like to use DEVONthink Pro Office for OCR, personally, but if you don't have that then you previously had to jump through a couple of extra hoops.)
[1]: http://www.macworld.com/2007/10/secrets/nov07geekfactor/index.php
[2]: http://www.macworld.com/article/131313/2007/12/scansnaps510m.html