Experiments have been performed to study the bremsstrahlung from 5 keV electrons incident on BeO and NaCl targets. The main objectives of the study were to compare the experimental results to those produced using pyPENELOPE (which is based on the main program of the Monte Carlo code, PENELOPE) in order to examine the accuracy of the code's implementation of the additivity approximation and its predictions for experiments involving insulating targets. Comparisons of experimental results to results produced using pyPENELOPE were made both for situations where surface-charging effects were significant and for situations where the insulating targets' surfaces were continuously neutralized. The results suggest that the PENELOPE code's main program, PENEPMA, does not simulate charging effects. Instead, it accurately simulates bremsstrahlung produced by low-energy electrons incident on insulating targets for situations where surface-charging is not significant. The results also support the validity of the additivity approximation, as incorporated into PENELOPE.