Devices using an electric field to produce flow in microchannel networks have application to precise chemical reaction, analysis and separation. There is a need for accurate computational design tools that can be used with the physically and geometrically complex conditions of practical devices. The equations governing electrokinetic reacting flow are presented together with classical one-dimensional cases that are directly relevant to the flows in electrokinetic devices. This provides the background for an order of magnitude study of the importance of the various terms in each governing equation, both for the conditions of the double-layer flow and for the main flow in the channel outside the double-layer regions. In agreement with previous studies, for channel widths in the range from several microns to hundreds of microns, it is found that representing the double layer using a local one-dimensional solution to produce the boundary conditions at the walls for the main flow is a good approximation. The 'layer model' that emerges is consistent with models proposed in previous studies under more restricted conditions than those considered here, where the role of non-uniform ion species concentration is analysed. The model is applied to the example of alternating reacting flow in a tee junction, both for a two-dimensional and a three-dimensional channel section. The case of the same flow driven by pressure instead of electric field is computed for comparison.