Ecosystem-based fisheries management (EBFM) is gaining traction worldwide, including in the Gulf of Mexico (GOM). Ecosystem models, such as applications of the Atlantis and Ecospace modeling approaches, are key tools for assisting EBFM. Patterns of spatial overlap between exploited fish species, other species of concern such as marine mammals and sea turtles, and human activities can have a large influence on the predictions made by ecosystem models, but these patterns are usually not well defined. We developed methods for producing distribution maps for the cetacean, sirenian, and sea turtle groups represented in the Atlantis model of the GOM, and employed a method, initially designed for fish and invertebrates, for generating preference functions for the dolphin species represented in the Ecospace model of the West Florida Shelf. Preference functions specify the preferences of species for certain environmental conditions and are used by Ecospace to allocate species biomasses in space. We also took advantage of our mapping outputs to estimate the percentage of spatial overlap between the hotspots of cetaceans and sea turtles in the US GOM and their areas of bycatch in the US pelagic longline fishery. The present study provides new insights into the spatial distribution patterns of marine mammals and sea turtles in the GOM large marine ecosystem, including the first quantitatively supported maps of Florida manatee (sirenian) distribution along the entire US GOM coast. Efforts such as ours should be continued for improving the reliability of ecosystem models and, thereby, advancing EBFM worldwide.