(RBBM)
I wouldn't consider the middle child's isotopes result areas as subjective, they're based on scientific data which are more like strong probabilities, IMHO. However, those results weren't restricted to the midwest, there were isolated, small areas in NH, VT and Northern Maine that also matched the results. Of these three areas, two would fit NCMEC's theory of her being with her father most of her life - which might mean the VT/NH areas.
He was also possibly in Quebec, which would lean more toward the Northern Maine connection.
You are certainly correct about one thing: NamUs doesn't have every missing person entered into their system.