Source code security has gained relevance in recent years due to the increase in software piracy. Several techniques have been developed to complicate the efforts of attackers who use reverse engineering to understand programs. Code obfuscation deliberately conceals its specifics by employing different approaches. This work evaluates the results of several studies that use one or more approaches. Based on this analysis, a software privacy model is presented, which is based on the diversification and obfuscation of the call graph. The model includes the tasks to be performed and the algorithms that implement it. Its evaluation was conducted in a testbed with favorable results.