Background: Owing to the exponential growth of three-dimensional (3D) environments amongst researchers and educators to create simulation games (SGs) in primary education, there is a growing interest to examine their potential support in computer science courses instead of visual programming environments. Objectives: This study explores the relationships between students' learning performance, focusing on computational thinking (CT) and programming skills development, in association with their cognitive load (mental load and mental effort) and emotions (happiness, anger, anxiety, sadness) by playing SGs. Methods: A total of two hundred and ninety participants (n = 290) in fifth-grade classes (10-11 years old) of Greek primary schools completed all pre-and post-intervention tests. A quasi-experimental study was conducted over a 14-week timetable course and in two comparison conditions, in which a SG was created by using OpenSimulator&Scratch4SL and Scratch for the experimental group (n = 145) and the control group (n = 145), respectively. To further investigate the effectiveness of the proposed SG in each teaching intervention, an exploration of relationships between students' learning performance, cognitive load, and emotions through multiple analyses, depending on correlation, t-tests, correlation, and hierarchal regression was delivered. Results and Conclusions: The findings indicate that the proposed SG created by using OpenSimulator&Scratch4SL positively affected students' emotions and cognitive load, whereas there was no significant difference in learning gain between the two groups. Implications: The study provides empirical evidence on the effects of SGs on students' knowledge acquisition, highlighting the importance of considering both cognitive and emotional components in the design of these games. The findings also offer valuable insights for implications policy and practice.