No, that doesn't make much sense. The point of a maximiser/limiter is to exploit the 0dBFS range as thoroughly as possible, but 0dBFS is only a meaningful range at the very end of the mastering process: it's the highest level that can be expressed by the consumer format you're exporting to. OTOH, digital plugins use floating-point arithmetic internally, which reaches to the absurd level of +1848 dBFS. Of course, that doesn't mean you should drive them so loud – for standard digital EQs it simply doesn't matter, and nonlinear / analogue plugins will sound differently depending on the input level, so you should drive them with a reasonable level that sounds good, likely something around -3 dBFS.
If you use a maximiser as simply an extreme compressor in front of another FX, the result will in general not be "maximised" any more. In particular, applying an EQ to a hard-maximised bass drum hit (the transients of which are effectively clipped by a limiter) will normally increase the level to something like +2 dBFS. So to get the result out for consumer, you will then need another limiter. Which would really harm the signal more than necessesary.
So: put the limiter at the very end of the chain. A reasonable setup might look something like softComp -> EQ -> tapeSaturation -> multibandComp -> maximiser/limiter
.