I am using a FM5 solution for document management by importing a referrence for scanned documents into a containor field. These cannot be used for "find" so the identifying data (used to find the document) needs to be entered manually. Can OCR'd scanned documents be entered into a text field that can then be indexed for "find"? If so, any suggestions on such a good scanner and OCR application. My Umax Astra MX3 is rather slow with 20 sec per scan and the OCR is useless.

Yes, once you OCR the document, you can put it in a text field, and Filemaker will index it and you can search for it.

I use Caere OmniPage Pro OCR software. It's quite accurate at conversion and will work automatically with a scanner that has an automatic document feeder so that you don't have to stand there and feed pages in one at a time. My scanner doesn't have the doc feeder option, so I can't say how well this feature works. It looks like you're going to be doing a lot of scanning, so buy a good scanner. I've had a Microtek scanner for about 5 years and have been very happy with it. They seem to make fairly good ones.