Photometric surveys of the Zodiacal Light (ZL) already allowed to retrieve features of interplanetary dust space distribution and optical behaviour. Of the brightness "gathering" function dZ = D(α)/ m along each line of sight (LOS), (α being the phase angle, m the subsolar distance of the LOS, D) the local scattering coefficient), two approximate values could be derived, based on the constraints provided by the two observed values of its integral Z, when the LOS is in-ecliptic. This "nodes of lesser uncertainty" method (Dumont, Levasseur-Regourd, Renard, 1985 to 1996), however, lowered but did not rule out question marks upon the phase function. To improve this inversion, additional constraints can be found in ZL surveys from deep space probes. We show that both the Pioneer 10 (Toller and Weinberg, 1985) and (despite their lack of in-ecliptic scans) the Helios (Leinert et al., 1982) data imply the phase function to weakly depart from isotropy, at least in the 30°-150° range. The latitudinal dependence f(β<SUB>⊙</SUB>, r = cst) of the space density (less well known than the heliocentric, D( r, β<SUB>⊙</SUB> = 0)) can be tracked through the brightness ratio, at the same elongation ɛ, aiming in the helioecliptic meridian, against in the ecliptic. At ɛ = 90°, this ratio 0.3 would lead—in the improper assumption of a single, homogeneous cloud—to fit the latitudinal density drop by a — cos<SUP>12</SUP>β<SUB>⊙</SUB> function. The resulting brightness ratio at ɛ < 90°, which should be equal to ∫<SUB>LOS</SUB> cos<SUP>12</SUP> β<SUB>⊙</SUB>(α)D(α)· dα/∫<SUB>LOS</SUB> D(α) dα turns out to be much lower than the ratio observed in the 60° > ɛ > 15° range (again ≃0.3). This contradiction is solved with a steeper exponent (20-22?) for cos/gb<SUB>⊙</SUB>, and by assuming the flattened cloud to coexist with another one, spherically symmetrical, which contributes 15-25 S10 at ɛ = 90°, 50-80 S10 at ɛ = 60°, 100-160 S10 at ɛ = 45° and 250-450 S10 at ɛ = 30°.< Réduire