Studies about effects of school lunch on children’s cognition are rare; two previous studies (CogniDo, CogniDo PLUS) generally found no negative effects of lunch on children’s cognitive performance at the end of lunch break (i.e. 45 min after finishing lunch), but suggested potential beneficial effects for single parameters. Therefore, the present study investigated the hypothesis of potential positive effects of school lunch on cognitive performance at early afternoon (90 min after finishing lunch). A randomised, cross-over intervention trial was conducted at a comprehensive school with fifth and sixth grade students. Participants were randomised into two groups: On day 1, group 1 did not eat lunch, whereas group 2 received lunch ad libitum. On day 2 (1 week later), group 2 did not eat lunch and group 1 received lunch ad libitum. The cognitive parameters task switching, working memory updating and alertness were tested using a computerised test battery 90 min after finishing the meal. Of the 204 recruited children, fifty were excluded because of deviations from the study protocol or absence on one of the 2 test days, which resulted in 154 participants. Data showed no significant effects of lunch on task switching, working memory updating and alertness (P values between 0·07 and 0·79). The present study suggests that school lunch does not seem to have beneficial effects on children’s cognitive functions regarding the conducted tests at early afternoon. Together with our previous studies, we conclude that school lunch in general has no negative effects on cognitive performance in children. However, beneficial effects seem to be restricted to a relatively short time period after eating lunch.