The ability to learn in a self-regulated way is important for adolescents' academic achievements. Monitoring one's own learning is a prerequisite skill for successful self-regulated learning. However, accurate monitoring has been found to be difficult for adolescents, especially for learning problem-solving tasks such as can be found in math and biology. This study investigated whether a self-explaining strategy, which has been found effective for improving monitoring accuracy in learning from text, can improve monitoring and regulation-choice effectiveness, and problem-solving performance in secondary biology education. In 2 experiments, one half of the participants learned to solve biology problems by studying video-modeling examples, and the other one half learned by giving step-by-step self-explanations following the video-modeling examples (Experiment 1) or by following the posttest problem-solving tasks (Experiment 2). Results showed that in contrast to earlier studies, self-explaining did not improve monitoring and regulation-choice effectiveness. However, the quality of self-explanations was found to be related to monitoring accuracy and performance. Interestingly, the complexity of the problem-solving tasks affected monitoring and regulation-choice effectiveness, and problem-solving performance. These results are discussed in relation to the cognitive demands that monitoring and regulating learning to solve problems combined with self-explaining pose on learners.