Once a document has passed OCR, it can not run through OCR again in Acrobat (that's an Acrobat limitation).
So, your options would be either start again from the original scans (preferrable), or export as TIFF and run OCR again (second choice, because there would be a second downsampling of the images).