Human enterovirus 71 (EV71) has emerged as a neuroinvasive virus that is responsible for several outbreaks in the Asia-Pacific region over the past 15 years. Appropriate animal models are needed to understand EV71 neuropathogenesis better and to facilitate the development of effective vaccines and drugs. Non-human primate models have been used to characterize and evaluate the neurovirulence of EV71 after the early outbreaks in late 1990s. However, these models were not suitable for assessing the neurovirulence level of the virus and were associated with ethical and economic difficulties in terms of broad application. Several strategies have been applied to develop mouse models of EV71 infection, including strategies that employ virus adaption and immunodeficient hosts. Although these mouse models do not closely mimic human disease, they have been applied to determine the pathogenesis of and treatment and prevention of the disease. EV71 receptor-transgenic mouse models have recently been developed and have significantly advanced our understanding of the biological features of the virus and the host-parasite interactions. Overall, each of these models has advantages and disadvantages, and these models are differentially suited for studies of EV71 pathogenesis and/or the pre-clinical testing of antiviral drugs and vaccines. In this paper, we review the characteristics, applications and limitation of these EV71 animal models, including non-human primate and mouse models.