I think that the MSJC (ACI530) actually contradicts itself, a little, with regard to biaxial bending of reinforced masonry.
As you pointed out, Section 2.2.3 (unreinforced masonry) points to the use of the Unity Equation. However, the commentary of Section 2.3.4.2.2 (reinforced masonry) explicitly says,
The interaction equation used in Section 2.2.3 is not applicable for reinforced masonry and is therefore not included in Section 2.3.
So it seems that the omission of the unity equation from Section 2.3 is intentional. Further, it isn't clear what that statement means, exactly. Is it implying that the equation explicitly written in Section 2.2.3 ( $\frac{f_a}{F_a} + \frac{f_b}{F_b} \leq 1$ ) is not applicable to reinforced masonry, or that the Unity Equation in general is not applicable?
However, I think that the language provided in Section 2.3.4.2.2 is fairly clear, or at least clear enough, to indicate how one might approach biaxial bending or (biaxial) bending and compression. It states,
The compressive stress in masonry due to flexure or due to flexure in combination with axial load, shall not exceed 0.45$f'_m$ provided that the axial compressive stress due to axial load component, $f_a$, does not exceed the allowable stress, $F_a$, in Section 2.2.3.1.
This seems to imply that, for axial force with bending, one would need to satisfy both,
$$\frac{f_a}{F_a} \leq 1 \tag{1}$$
and,
$$\frac{f_a + f_b}{0.45f'_m} \leq 1 \tag{2}$$.
This would then lead me to rationally assume that for pure biaxial bending we'd simply need to satisfy,
$$\frac{f_b^1 + f_b^2}{0.45f'_m} \leq 1 \tag{3}$$
which is just a form of the Unity Equation. Hence, the code seems to contradict itself, kinda.
A few final things worth noting:
- I have a couple references, written to ACI380-11, that qualify masonry beams for biaxial bending using equation 3 (published by PPI),
- I see all sorts of stuff online, from reputable schools and organizations, using equation 3 to qualify reinforced masonry beams for biaxial bending,
- Don't forget that you need to check the combined stress in your steel as well. This conversation has largely focused on the stress in the masonry, but ensure that
$$\frac{f_s^1 + f_s^2}{F_s} \leq 1 \tag{4}$$