In the present paper, the authors have applied a newly developed semi-empirical, zero-dimensional, two-zone model on two automotive DI Diesel engines: a heavy (truck) and a light-duty (passenger car) engine. The aim of the study is to examine model's ability to predict NOx trends with the variation of engine load/speed, fuel injection timing, EGR rate, boost pressure and fuel injection pressure. The model makes use of cylinder geometrical data, basis engine operating data and the experimental combustion rate, as deduced from heat release analysis, to calculate engine tailpipe exhaust NO emissions and their formation history inside the combustion chamber. Apparently, combustion and hence NOx formation, occurs inside the burnt zone. The required air for the combustion is provided from the unburnt zone. The entrained air mass, at each time step, is calculated using a mean equivalence ratio value and the corresponding fuel mass burnt during the calculation time step (as defined by the experimental combustion rate). The mean equivalence ratio value, at each operating point, is obtained using a new correlation which makes use of parameters derived from the processing of the measured pressure trace, as well as typical engine operating parameters. NO (the predominant part of total NOx) is calculated using the extended Zeldovich mechanism at each time step (during combustion and expansion). Thus, the NO formation history inside the cylinder is provided, considering the two zone approach. The specific concept of the two-zone approach combined with the utilization of engine's measured parameters, the empirical correlation for the mean equivalent ratio and the simplicity of the model's calculations, provide a robust and time efficient tool for NO prediction which can be applied on various engine configurations and operating conditions without extended calibration. The derived results reveal model's ability to predict exhaust NO emissions and trends satisfactorily at a variety of engine settings. Considering its low computational cost, it can serve as a useful tool for emissions trade-off optimization studies, as well as for real-time model based NOx control applications.