Merging without re-rendering
There are two ways to combine PDFs. The lazy way renders each page to an image and builds a new document from the pictures — fast to implement, and it destroys every useful property the file had: text stops being selectable, file size balloons, and printing quality drops.
This copies page objects between documents instead, carrying their fonts, embedded images, vector artwork and annotations with them. The merged file is the sum of its inputs, not a photograph of them.
Order is the whole job
Most merges are assembling something: a contract with its appendices, a report with its charts, scanned pages that came out of the feeder in batches. Getting the sequence right matters more than anything else, which is why the list is reorderable before you commit rather than after.
To pull a document apart instead, use Split PDF. To rotate or drop individual pages, use Organize PDF. If the result is larger than you need, Compress PDF reports honestly what it can and cannot save.