That’s the point to Document Understanding. You use Document Classification to determine which pages are which documents, then you can use the classification output to get the page ranges and split them out to separate PDFs with the Extract PDF Page Range activity.