I want to study the turbulence inside an isothermal reactor. What is the better model I can use? Until now I have used only k-epsilon model.Is it correct this model for such study or is better another model?

Each model has its own advantages and disadvantages so the choice of model is up to the user based on what flow characteristics you want to capture. K-epsilon models are good for modeling the freestream but tend to lose their accuracy in the near wall simulation, whereas k-omega models are the opposite. The SST formulation of the k-w model tries to blend k-w near the wall and k-e near the freestream. It has the disadvantage of over-predicting the onset and reattachment of flow separation, though. There's a great deal of literature available from Fluent on the theory behind the difference turbulence models.