The present work evaluates several mixing models for the prediction of premixed combustion in a Partially Stirred Reactor (PaSR). The models considered in this work were the extended IEM, Langevin and extended Langevin models. The degree of mixing and its influence on the termochemical properties in a combustion process are investigated here. The first part of this work consists on the presentation and the assesment of these mixing models in which a single scalar field was considered in presence of a homogeneous and isotropic turbulent field. Since these mixing models involve stochastic terms, their implementation is performed by the Monte Carlo method using numerical schemes which solve the corresponding Stochastic Differential Equations (SDE). The evolution of the Probability Density Function (PDF) and the main properties for a single scalar field are studied for each mixing model. The numerical results are compared with Direct Numerical Simulation and available analytical results. Excellent qualitative and quantitative agreements are obtained. In the second part of this work, mixing models are used for numerical simulation of a PaSR where the diffusive and reactive processes occur. The PaSR is used to assess the mixing model influence on the termochemical properties of the mixture in a premixed combustion process, which is modeled using a reaction progress variable. The results obtained with the different mixing models are compared in several operating regimes of the PaSR, showing that when mixing is fast and reaction is intense, the different models lead to similar results. However, when mixing is slow and reaction is weak, important discrepancies are observed between the model results, which reach 65%, as far as the averaged reaction progress variable is concerned