The size is something of an exagguration, as is the location in the Bahamas, but otherwise the idea of a maroon village in the interior of one of the larger West Indian landmasses is fairly reasonable.

There's been quite a lot of recent archeological work on settlements of escaped slaves and their descendents, like this recent article in the Smithsonian about maroon settlements in the swamps of Virginia and North Carolina.

Most maroon settlemements, however, were located in the forested and mountainous interiors of the larger Caribbean islands like Jamaica, Hispanola, and Puerto Rico. There were also maroon settlements on the Central American and South American mainland, where such settlements had a great deal of cultural and economic exchange with the Native commuinities.

The maroon settlements of Jamaica are probably the best-attested with respect to both both archeological evidence and written historical records. Estimates of settlement population have varied widely, but most maroon villages seem to have been quite small, perhaps 50 persons or fewer, although newer work has tended to say that archologists underestimated the number of persons per household and that settlements of 100-300 indivudals were not extant by the early to mid 18th century.

Alvin O. Thompson's Flight to Freedom: African Runaways and Maroons in the Americas is my source for this answer, although it's way outside my area of specialty and I honestly just read it for my own curiosity.