We present a subgrid-scale (SGS) scalar variance model based on the explicit algebraic subgrid scalar flux model, EASSFM (8). The EASSFM is a dynamic mixed nonlinear tensor eddy diffusivity model, which is derived from the modeled transport equation of the SGS scalar flux. The explicit algebraic form is obtained using the weak equilibrium assumption. The resulting model improves the direction of the predicted SGS flux vector and enables the prediction of shear-induced SGS fluxes, in contrast with the eddy diffusivity model. The EASSFM has been used for large eddy simulation (LES) of turbulent channel flow with and without system rotation (8; 9) and has been found to improve LES predictions over the dynamic eddy diffusivity model. A priori analysis of the EASSFM using the filtered DNS data from a reacting turbulent wall-jet has been performed recently (6; 7), which also showed favorable results. In this study, we evaluate our SGS scalar variance model using the filtered DNS database of a turbulent reacting wall-jet, which is an extension of our previous study on reactive turbulent wall-jet flows (5; 7) to a larger simulation domain. The results show a good agreement between the filtered DNS and our model predictions for the passive and active scalars. This indicates that acceptable predictions of the SGS scalar variance can be obtained using the EASSFM with the new SGS scalar variance model.