In this paper a geometric approach to the special relativity (SR) is used that is called the “invariant special relativity” (ISR). In the ISR it is considered that in the four-dimensional (4D) spacetime physical laws are geometric, coordinate-free relationships between the 4D geometric, coordinate-free quantities. It is mathematically proved that in the ISR the electric and magnetic fields are properly defined vectors on the 4D spacetime. According to the first proof the dimension of a vector field is mathematically determined by the dimension of its domain. Since the electric and magnetic fields are defined on the 4D spacetime they are properly defined 4D vectors, the 4D geometric quantities (GQs). As shown in an axiomatic geometric formulation of electromagnetism with only one axiom, the field equation for the bivector field F [33], the primary quantity for the whole electromagnetism is the bivector field F. The electric and magnetic fields 4D vectors E and B are determined in a mathematically correct way in terms of F and the 4D velocity vector v of the observer who measures E and B fields. Furthermore, the proofs are presented that under the mathematically correct Lorentz transformations, which are first derived by Minkowski and reinvented and generalized in terms of 4D GQs, e.g., in [23], the electric field 4D vector transforms as any other 4D vector transforms, i.e., again to the electric field 4D vector; there is no mixing with the magnetic field 4D vector B, as in the usual transformations (UT) of the 3D fields. Different derivations of these UT of the 3D fields are discussed and objected from the ISR viewpoint. The electromagnetic field of a point charge in uniform motion is considered and it is explicitly shown that 1) the primary quantity is the bivector F and 2) that the observer dependent 4D vectors E and B correctly describe both the electric and magnetic fields for all relatively moving inertial observers and for all bases chosen by them. This formulation with the 4D GQs is in a true agreement, independent of the chosen inertial reference frame and of the chosen system of coordinates in it, with experiments in electromagnetism, e.g., the motional emf. It is shown that the theory with the 4D fields is always in agreement with the principle of relativity, whereas it is not the case with the usual approach with the 3D quantities and their UT.