Several studies indicate that executive functions (EF), such as working memory (WM), inhibition or flexibility can be improved by training and that these training-related benefits in WM capacity generalize to reading and mathematical abilities. However, the results of these studies are inconsistent and most of them focused on WM training in children with learning difficulties. Evidence for typically developing children is rare and no study has investigated inhibition training or flexibility training. There is also a lack of studies taking motivational factors into account. Therefore, this study compared the effects of game-based and standard training regimens targeting WM, inhibition, or flexibility in children. One hundred and fifty-three typically developing elementary school students (mean age = 9.6 years, standard deviations = 0.8) were investigated in an intervention design with a pretest, 21 sessions of training, a posttest and a follow-up after three months. They were randomized into one of six training groups or a control group. We found training gains in all training groups and higher self-reported motivation in the game-based as compared to the standard training groups. Furthermore, there was domain-specific transfer to untrained EF tasks across all training groups. We found greater performance improvements in reading ability (but not mathematics) in the game-based flexibility training group and the game-based inhibition training group as compared to the control group. Transfer effects were still significant at follow-up. In sum, our findings provide first evidence for a systematic comparison of training on different domains of EF and their differential effects on academic abilities.