The high priority of military training programs in the army is to improve the resilience of its professional and reserve soldiers. In addition, resilience helps to maintain the optimal cognitive and physical performances necessary for mission success; it also has other positive effects that have not yet been studied. Therefore, the purpose of the current study is to understand the effect of psychological resilience (PRE) on the intention of reservists to remain in the active reserve (FMT) after the completion of combat training. In addition, we focus on the motivation (prosocial motivation (PRM) and intrinsic motivation (IMT)) and psychological capital (passion in the pursuit of long-term goals (LTGs), proactivity (PRO), and capability to harness the circumstances (SEF)), and achieved a level of competence after the completion of training (ACH). Building on the perspective of training sustainable reserve soldiers’ capabilities and the theory of psychological resilience, we hypothesize and assess the effect of psychological resilience on the intention of reservists to remain in the active reserve for future military training, as well as the mediating role of psychological capital, motivation, and the obtained training ratings. The study hypotheses are tested on a sample of 345 Lithuanian reserve soldiers. The mediating effects are tested by employing the PROCESS v3.5 macro program developed by Hayes, and bootstrap 10,000. The main findings propose that the total impact of psychological resilience is highly positive (β = 0.659, t = 7.670, p < 0.001) and influences the decision of the reservists to extend their capabilities in future military training. Furthermore, intrinsic motivation (β = 0.691, p < 0.001, and 58.6% of the total effect) and the achieved rating of completed training (β = 0.136, p < 0.01, and 21.2% of the total effect) positively mediate the relationship between psychological resilience and the intention to remain in the active reserve. Selected bootstrap confidence interval and Sobel test statistics with two-tailed probability values evidence the robustness of the mediating results. The significance of the theoretical and practical implications is discussed.