The estimation of electrical power generation in photovoltaic (PV) grid-connected systems based on meteorological data is a nontrivial, highly useful task, for instance to achieve accurate energy assessment. Widely used PV generation simulators are PV Systems (PVsyst), System Advisor Model (SAM) and PVLib. These simulators are characterized by presenting numerous features and providing complete results, however the PV estimation model SoL is an example of a new approach to PV generation estimation. SoL is characterized by its simplicity and computational efficiency. The objective of this paper is validating the recently published SoL model using real data from two PV locations for several years and facilities and comparing the results with those of three other PV simulators, namely PVsyst (in Spain), SAM (in Denver) and PVLib (both). It has been found that SoL estimates power production accurately for both locations and its estimations are more precise than those given by PVsyst, SAM and PVLib. It proves to be more computationally efficient than PVsyst, it can work with higher resolutions than SAM and PVsyst and requires fewer inputs than PVLib, SAM or PVsyst. Finally, a self-shading model is proposed as an enhancement for the SoL model. The number of inputs required is minimal, and it is an approximate yet efficient model. The estimation when using the self-shading enhancement is even more accurate than the previous estimation for SoL in locations where self-shading is evident. SoL proves to be an appropriate model for power estimation, and its results are enhanced when using the self-shading model proposed in this paper.