Complex electrical impedance has real and imaginary part for resistance and reactance respectively. Modulus is usually expressed in ohms. What physical information is conveyed from modulus alone without knowing complex parts?
This is an interesting and difficult question. A way to approximate this physical meaning is, for example, to think that if you are driving a RL circuit with direct current, the effective impedance modulus │Z │value coincides with the resistance value; but, with alternating current, the impedance modulus attains a bigger value, indicating that part of the applied voltage is lost in the reactive part. Only, if you add the Z angle information to the modulus value, the inductive or capacitive value of the impedance could be determined. The impedance modulus integrates the effective combined opposition to the pass of an electrical current through a circuit due to its internal resistive and reactive character for each specific applied frequency.
From a more general point of view, the impedance modulus represents the voltage amplitude for a current amplitude of value "unit" flowing through a circuit.
If by modulus of electrical impedance you mean the absolute value of impedance
|Z|=Sqrt [ Z1(w)^2+Z2(w)^2 ],
then nothing new is learned from this number that can not be learned from the real (Z1(w)) and imaginary part (Z2(w)). It is just another way to represent a complex number (in phase and out of phase part of the electrical response) and without phase angle, as pointed out by Antonio and Juan, you have only part of the response information. So , stick to the real and imaginary part of the impedance, all information is there ! You can of course display the given response information by performing some trivial transforms which brings you to electrical complex admittance Y (Y(w)=1/Z(w)) or to the complex electrical capacitance C (C(w)=Y(w)/i.w . There are some other quantities that are unfortunately often used, such as "complex dielectric constant" (term dielectric function should be used !) or electrical conductivity, but these quantities are often un-physical (they do not represent what they should, namely the two intensive material parameters) if a clear distinction is not made between bulk and near interface regions that are being measured.
one interpretation to attach a physical meaning may perhaps be obtained from DSP
filter designs where often only the magnitude of the transfer function is used to
obtain relationships of input and output magnitudes. In circuit analysis this means
e.g vm sin(wt) applied to z= R+jwL results in a current of magnitude im =vm/mod(z). In an instantaneous plot the peak of voltage and peak of current relation only will be known
not the relative location, further this would only be applicable for steady state so