We analyze in detail a mathematical model of capillary zone electrophoresis (CZE) based on the conception of eigenmobilities, which are eigenvalues of the matrix tied to the linearized continuity equations. Our model considers CZE systems, where constituents are weak electrolytes and where pH of the background electrolyte may reach the full range from 0 to 14. Both hydrogen and hydroxide ions are taken into account in relations for conductivity and electroneutrality. An electrophoretic system with N constituents has N eigenmobilities. We reveal that two of the eigenmobilities have a special meaning as they exist due to the presence of hydrogen ions and hydroxide ions (in water solutions). These two eigenmobilities are responsible for the existence of two corresponding system zones (system peaks). We show that the stationary zone (injection zone, water zone, gap, peak, dip) is in many common background electrolytes composed of these two eigenzones which overlap, due to their very low electrophoretic mobility, into one zone. Other eigenmobilities give rise to system zones originating due to a possible existence of double (or multiple) coconstituents in the background electrolyte. The last group of eigenmobilities is connected with the movement of eigenzones accompanying analytes and enabling their indirect UV or conductivity detection. The model allows assessing experimentally available quantities such as effective mobility of the analyte, molar conductivity detection response, transfer ratio, and relative velocity slope and gives a picture about migration of analytes, their electromigration dispersion and signals obtained in detectors. It allows computer simulation of electropherograms and enables optimization of background electrolytes.