Black holes generally are defined as stellar objects which do not release any light. The Schwarzschild radius, derivedfrom GRT, defines the horizon radius for non-rotating black holes. The Kerr metric is supposed to define the “eventhorizon” of rotating black holes, and this metric is derived from generally “acceptable” principles. The limit for theKerr metric's horizon for non-rotating black holes is the Schwarzschild radius.By analyzing the horizon outcome for rotating and non-rotating black holes, using the Maxwell Analogy for Gravitation(MAG) (or historically more correctly: the Heaviside Analogy for Gravitation, often called gravitomagnetism), Ifind that the Kerr metric must be incomplete in relation to the definition of “event” horizons of rotating black holes. Ifthe Maxwell Analogy for Gravitation (gravitomagnetism) is supposed to be “a good approach” of GRT, we may assumethat it is a valid analysis tool for the star horizon metrics.The Kerr metric only defines the horizons for light, but not the “mass-horizons”. I find both the “light-horizons” and thethe “mass-horizons” based on MAG. Moreover, I deduct the equatorial radii of rotating black holes. The probable originof the minutes-lasting gamma bursts near black holes is unveiled as well. Finally, I deduct the spin velocity of blackholes with a 'Critical Compression Radius'.The deductions are based on the findings of my papers “Did Einstein cheat?”, “On the geometry of rotary stars andblack holes” and “On the orbital velocities nearby rotary stars and black holes”.