Why the generic page exists
The format-specific pages assume you know what you have. In practice a folder that needs to become a document holds a mix: a couple of photos from a phone, a screenshot, something exported from a design tool. Sorting them by format first, converting each group separately and then merging is three steps too many.
So this page lets you convert image to pdf from everything the image engine can decode, mixed freely. Each file is decoded, scaled if it is enormous, and written as one page. The output is one document whatever went in.
HEIC, which is the one that usually fails elsewhere
Since iOS 11 an iPhone saves photos as .heic rather than .jpg, and no browser can display one. Most image to PDF converters therefore reject them, which is particularly unhelpful given that photos taken on a phone are exactly what people want to turn into a document: receipts, forms, a passport page, a signed agreement.
These are handled here through the same decoder the HEIC conversion pages use, so a camera roll goes in without a preparation step. The one visible difference is that HEIC rows show a file icon rather than a thumbnail, because drawing a preview would mean a full decode for a 40 pixel image.
Choosing page size and margin
Fit-to-image gives every page the shape of its own image. Nothing is cropped, no white bands appear, and the file stays as small as it can be. This is the right default for a document that will be read on a screen.
A4 and US Letter give every page a standard printable size, centre the image and rotate to landscape where the image is wider than tall. Choose one of these when the document is going to a printer or into a system that expects paper sizes, and add a margin so a printer that cannot reach the edge of the sheet does not clip anything.
Everything runs in your browser, and nothing is sent to a server.