Terry, you are absolutely correct, on both counts.
the hash from an ST5 is identical to the hash from an FFP. both make the same checksum from just the audio data, ignoring tags (and ignoring FLAC version and ignoring compression level. i.e. you could take a FLAC v1.1.1 (level 5) and re-encode to FLAC v1.3.0 (Level

and the FFP would not change, because the audio data hasn't changed).
and the FFP verify tool in TLH does indeed check that the PCM audio is bit-for-bit accurate with the original, ignoring encoding/metadata/tags. that's why it doesn't happen instantaneously. when you make an FFP in TLH (or any other program), it happens in a blink of an eye, because it's just grabbing the already calculated FFP that is stored in the file header. but when you verify the FFP, it actually re-generates the FFP and compares it to the FFP that is stored in the file header.
because of all that, ST5's are generally redundant and not needed. the only time I would recommend using an ST5 is when you are converting an old SHN file set to FLAC. Run the ST5 on the original SHN files. then create your FFP after you convert to FLAC. you can now compare the SHN ST5 to the FLAC FFP and confirm that no audio data was changed.