This is, I believe, a question that biologists are more suited to answer than historians, and you will notice that the paper was published in PLoS Biology.

That being established, the linked study does not make any claim on the number of people who originally settled the Americas, rather, it claims that the effective population of the colonizing group was less than 80 individuals. The effective size is an idealized population which perfectly follows certain criterion such as perfectly random mating, to simplify the maths.

The effective population, thus, is always smaller than the real population, as it is diminished by imperfections such as inbreeding and lack of equal chances of reproduction. Hence, it's quite likely that the real founding population was bigger, although by how much is hard to say.

Finally, the generally accepted criterion for long-term viability of a population is the effective population must be at all time greater than 50 individuals, and on the long term must be superior to 500. Given the ressource availability in the New World, this small population would have been able to expand considerably, and effective population could have bounced back above the 500 mark relatively rapidly, so the result is not implausible.

As for the language families, you will note the study has been exclusively conducted on Amerind-speaking populations. The Americas have been colonized multiple times, the latest group to do so being the Inuit. This study only apply to one such colonization event, other colonizating groups would have brought their own language families, as the Inuit did.