Recent advances in QEEG-databases have enabled improvements in interpretation, which in turn have led to more effective neurofeedback interventions. These improvements relate mostly to evaluations conducted in the high frequency beta range (21–30 Hz) evaluation and in single Hz bins, which more specifically address which frequencies need to be trained to most quickly and effectively normalize their dysfunctions and remediate their difficulties. Use of the modular activation/coherence model (Walker et al. in J Neurother 11: 25–44, 2007) provides a framework for correcting the slow or fast modular dysfunctions, as well as normalizing connectivity using coherence training.