I am sorry, I accidentally replied to an old post. I have no input for your current question.
I believe Adobe would rather people did not use Paper Capture through an API, and while there happens to be an API they have no intention of documenting it any further or making it more usable. I should say this sort of work is outside the intended scope of Acrobat, as an interactive user tool. There are, I am sure, products much more suitable for batch OCR of PDFs. From Adobe, I believe some of the LiveCycle products do this for a server environment.