This systematic review focuses on three themes: 1) the nature of pre-sleep cognitive activity in good sleepers and individuals with insomnia, 2) the links between measures of pre-sleep cognitive activity and sleep onset latency (SOL) or insomnia, and 3) the effect of manipulating pre-sleep cognitive activity on SOL or insomnia. Regarding the first theme, mentation reports have been collected in a sleep laboratory, with an ambulatory monitoring device, or using a voice-activated tape-recorder. Normal transition to sleep is characterized by sensorial imagery, deactivation of higher cognitive processes, and hallucinations. Moreover, pre-sleep thoughts in individuals with insomnia frequently relate to planning or problem-solving, and are more unpleasant than in good sleepers. Regarding the second theme, twelve questionnaires and three interviews were identified. Insomnia is associated with more thoughts interfering with sleep, counterfactual processing, worries, maladaptive thought control strategies, covert monitoring, and cognitive arousal. Regarding the third theme, several strategies have been tested: mental imagery, hypnosis, paradoxical intention, articulatory suppression, ordinary suppression, and distraction. Their effect is either beneficial, negligible, or detrimental. Future research should focus on the mechanisms through which some forms of cognitive activity affect sleep onset latency.