Nitrogen oxides (NO x) are one of the main harmful emissions from diesel engines. Regulations on emissions are becoming more stringent; consequently, research should aim at reducing both engine-out NO x and tail-pipe NO x emissions. Exhaust gas recirculation is mainly used to reduce engine-out NO x emissions. After-treatment methods such as lean NO x trap systems and selective catalyst reduction are used to minimize tail-pipe NO x emissions. Real-time feedback control during transient conditions can further reduce these emissions and be utilized when information about real-time engine-out NO x emissions is available. It would be helpful to evaluate the proportion of NO2 in NO x emissions because the conversion efficiency of an after-treatment system is affected by temperature and the NO-to-NO2 ratio. Therefore, a semi-empirical NO x model considering NO and NO2 separately was developed for a diesel engine in this study. The NO estimation model was established based on the extended Zeldovich mechanism. The NO2 estimation model focused on chemical reactions between NO and other species and was based on formation and decomposition mechanisms. This NO x model can contribute to the cost reduction of engine systems by replacing existing NO x sensors and can also be applied to real-time feedback control strategies to minimize NO x emissions.