The aim of this paper is to provide insights for better modeling hub location problems to help create a road map for future hub location research. We first present a taxonomy to provide a framework for the broad array of hub location models, and then seek to identify key gaps in the literature that provide opportunities for better models. We provide some new perspectives in several areas, including the historical evolution of hub location research, models for economies of scale, and relevant characteristics of different applications. We also provide a succinct summary of state-of-the-art formulation and solution approaches. We conclude with a set of themes that can be addressed in the future for better modeling hub location problems.