In order to understand the physiopathology of epilepsies and develop antiepileptic drugs, animal models have been developed. These models appear to be valuable predictors of treatment efficacy; however, several of the currently used models remain questionable and probably inappropriate for the search for new treatments, in particular for epilepsies that cannot be treated by current antiepileptic drugs. In the present review, we report the results of a recent survey conducted by neurologists in charge of an epilepsy programme based at different hospitals in France. The 36 experts were questioned, via the internet, on the most critical features of four prototypic forms of epilepsy (idiopathic generalised epilepsies with convulsive seizures, absence epilepsy, focal epilepsy associated with dysplasia, and focal epilepsy associated with hippocampal sclerosis) that should be taken into account with regards to the relevance of animal models of epilepsy. Their answers suggest that most current models for focal epilepsies associated with either dysplasia or hippocampal sclerosis do not address the most relevant features. The models currently used in mice and rats are discussed in light of the data obtained in our survey.