The EEG source localization is an interesting area of research because it provides a better understanding of brain physiology and pathologies. The source localization accuracy depends on the head model, the technique for solving the inverse problem, and the number of electrodes used for detecting the EEG. The purpose of this study is to examine the localization accuracy of the eLORETA method applied to high-density EEGs (HDEEGs). Starting from the 256-channel EEGs, three different configurations were extracted. They consist of 18, 64, and 173 electrodes. The comparison of the results obtained from the different configurations shows that an increasing number of electrodes improve eLORETA source localization accuracy. It is also proved that a few number of electrodes could be not sufficient to detect all active sources. Finally, some sources resulting significant when few electrodes are used could turn out to be less significant when EEG is detected by a greater number of sensors.