In this work, a unique model is proposed for predicting the tensile strength of binary polymer blends considering the effects of polymer/polymer interface and the morphological variation of the system. The modeling was performed based on the combination of analytical and artificial neural network (ANN) modeling methods. For the analytical part, Kolarik’s model was developed in accordance with the system requirements and ANN was simultaneously involved in order to interpret some effective model parameters using the tensile test result of an actual sample (e.g. the yield strength and thickness of the interface, etc.). Furthermore, the model accuracy was evaluated by comparing the tensile test results of differently prepared iPP/PA and PS/PMMA blend samples and also some other data from literature with the model predictions. It was revealed that the designed ANN perfectly elevates the capability of the analytical section in order to predict the tensile strength of binary polymer blends with different compositions (prediction error ＜ 10%).