This article aims to take a new look at the explanatory abilities of the cognitive resource approach. We argue that observed limitations are logical consequences of the normal functionality of the cognitive system. Those limitations allow us to maintain a consistent implicit theory of oneself. We assign a crucial role to the subjective perception of self-efficacy and expectation of success on the effectiveness of solving cognitive tasks. This approach emphasizes not only the upper border (one cannot perform better than their expectations) but also a lower border (one cannot perform significantly below their expectations). Using different experimental data provided by ourselves and various research groups, we demonstrate relations between one's self-efficacy evaluation and their real-life results.