OK but the thing about the 1st floor is that it's the most heavily trodden floor in the entire dungeon. Most of the thrill-seeking Waterdhavians who descend into Undermountain won't go beyond this first floor. It's been picked clean over and over again.
The second floor is much less empty. There's even a goblin market on that floor!
Logic may agree with you. Fun, exciting gameplay to capture the imagination of your players does not.
Go big. Get them out of the gate.
I'll give you an example from a megadungeon that I run for my home games - which I was writing for publication before 3.5 ended and the 4E GSL derailed my career writing for D&D.
The megadungeon complex is modelled after the Mammoth Cave system located near where I live. It sprawls beneath a frontier community where the locals have legends about the caves, but they've never delved deeply enough to encounter the other humanoids who live down there. You have interactions such as the chimney of an old forge, which is poking out of the top of a knob. The teenagers from the village dare themselves to jump across it. One night someone falls in and the heroes have to go down there to recover the kid.
Or you have the haunted plateau where an elemental spirit keeps trying to cover an entrance to the dungeon with sand. This is just a simple unseen servant, but to the locals, it's a terrifying apparition.
The known entrances show graffiti from generations of spelunkers. There are dangerous chasms and ancient traps that have kept the other residents from getting farther.
Sure, this method is a slow burn, but at least there is something interesting going on. It takes a few rooms to begin to see something happened to the first level's civilization. With Dungeon of the Mad Mage, there isn't even anything clear about "should we explore deeper?" It's just nothing. No inspiration.
I'm not going to belittle the cartography, but the dungeon is too large physically for the scope of its ideas.