Brant Gardner, in his book "Traditions of the Fathers," argues that for a geographic correlation to be compelling, it must be productive. In other words, the correlation feeds back into the text expanding our understanding of it and explaining some other oddities.
One example he highlights is the Lamanite attack on Ammonihah. In the text, this is the only Lamanite attack (or one of the only) that does not put the subject city under tribute. Instead, it's a lightning raid that carries off captives. It's an oddity in the text. But place that story in a Mesoamerican context, and that context expands our understanding of what happened. The Lamanites have a new king. A coronation ceremony requires human sacrifices taken in battle. The Anti-Nephi-Lehis aren't fighting back and so aren't suitable for this purpose. Thus, the raid on Ammonihah, the emphasis on captives, and the lack of an actual conquest of the city.
I subscribe to the Sorenson model, but not just because the book fits inside that cultural region, but because the cultural region fits inside the book.