[en] The computer code, Stopping and Range of Ions in Matter (SRIM), is widely used to describe energetic processes of ion-solid interactions; its predictive power relies on the accuracy of energy loss/transfer and collision processes being considered. While the SRIM code is commonly applied in radiation effects research to predict damage production and in the semiconductor industry to estimate ion range and dopant concentration profiles, two challenges exist that affect its use: (1) inconsistency in estimations of atomic displacements between full-cascade and quick (modified Kinchin–Pease) options and (2) overestimation of electronic stopping power for slow heavy ions in light targets (e.g., Be and Si) or in compound targets containing light elements (e.g., C, N and O in carbides, nitrides and oxides). Based on a literature review and our experimental investigations, we discuss the underlying reasons for the discrepancies, clarify the physical limitations of the SRIM predictions, and, more importantly, provide recommendations to address the two challenges.