Parkinson's disease(PD) is a chronic, progressive neurodegenerative disease that primarily affects the motor function of the central nervous system. Due to its complex pathogenesis and lack of effective treatments, the study of PD has been one of the hot topics in the field of neuroscience. In recent years, with the continuous progress of science and technology, remarkable progress has been made in the study of animal models of PD. The neurotoxin model showed obvious damage to dopaminergic neurons, which was mostly used for the screening of neuroprotective drugs, but no pathological features were found. Transgenic models are beneficial for the discovery of potential therapeutic targets, but no significant dopamine neurodegeneration has been observed. The combination of neurotoxins and transgenic models can overcome their pathological shortcomings, resulting in more motor phenotypes and non-motor symptoms. α-synuclein-inducible models can show significant aggregation of pathological proteins. Cell models can be reproduced in vitro for the reproducibility of pathological features and used as pre-screening for animal models of drugs. In recent years, with the popularity of the Braak hypothesis, more and more studies have shown that PD is more likely to be a syndrome, and animal models of non-motor symptoms of PD are expected to identify effective biomarkers in the pre-disease stage and provide treatments for the most disabling non-motor symptoms. In this review, we will review the advantages and limitations of various animal models, aiming to provide a reference for researchers to select experimental models based on experimental purposes, and to provide a reference for an in-depth understanding of disease pathogenesis and the development of new treatments.