Among the multiple possibilities to study human pathologies, animal models remain one of the most used pathways. They allow to access to unavailable answers in human patients and to learn about mechanisms of action of drugs. Primarily developed with rats, animal models in anxiety have been adapted with a mixed success for mice, an easy-to-use mammal with better genetic possibilities than rats. In this review, we have focused on the most used animal models in anxiety in mice. Both conditioned and unconditioned models are described, to represent all types of animal models of anxiety. Behavioural studies require strong care for variable parameters, linked to environment, handling or paradigm; we have discussed about this topic. Finally, we focused on the consequences of re-exposure to the apparatus. Test-retest procedures can bring in new answers, but should be deeply studied, to revalidate the whole paradigm as an animal model of anxiety.