This should be a somewhat shorter post than normal.
Progressive / mid-game scoring requirements are good. They require you to proactively solve problems on the board throughout the game, instead of focusing on simply structuring late-game grabs. This diversifies list design in a positive way.
Additionally, this encourages engagement and CAN (if executed correctly) broaden the value of more units ... after all, if the game is something other than "camp wherever you need with super unit, then snag late with hidden throwaways" that's probably a good thing for those units that don't fill one of those two categories.
Several folks have been modifying Maelstrom also in some basic ways to get rid of the rando d3 value components, and to allow for immediate discard of impossible objectives, etc. etc. This helps as well.
BUT it's not all good, unfortunately ...
Where I struggle with Maelstrom, both conceptually and from having played with them, is that they turn the "Maelstrom" component of the game into the picture that started this article.
They basically give you an assignment, and tell you to go accomplish it as efficiently as you possibly can, so as to continue drawing new assignments. It's like power questing in MMOs, written into 40K. There's a certain element of flexibility required in accomplishing them, and you need to optimize your list for the many variables for accomplishment that are involved. That said, it doesn't make accomplishing them a tactically or strategically meaningful accomplishment. At least personally, I find it diminishes the feel of the game when the game itself says "immediately go do this thing, right now."
Let's compare this to a situation where you have multiple fashions in which to progressively score points during a game (i.e., objectives across the board are worth points if held). In a situation like that, the game is still telling you "collect points during play!" but it is not telling you "collect points during play by going to this exact coordinate on the board as quickly as you can."
To help draw the comparison, imagine if instead of Kill Points or Points Destroyed, you randomized your opponent's army into #'s based upon each unit, wrote those numbers sequentially on a deck, shuffled it, and then drew randomly from the top. You would only gain points by killing the current kill-assignment. Now, instead of figuring out the target priority and risks that will best win you the game, you literally just do what the cards tell you in each and every moment. It's like the game of 40k is playing your models for you.
The difference is perhaps subtle, but it's enormous. In the case of Maelstrom, the game is basically telling you what your tactics and strategy are. In the case of more player-choice progressive scoring, you're selecting from the myriad of possible strategic courses which one you think will best accomplish a net win in the end. Both approaches impact list design in very similar ways, and both of them are better than a simple "grab points at the end and who cares what you do until then" environment. By being random, specific, and force-targeted, however, Maelstrom still pales next to the gaming situation created by more player-choice multi-route progressive-scoring mission design.
So we're halfway there. We'll see if we can't get all the way there while echoing some of the soul of the Maelstrom with our heavily revised mission packet, due out by July 1 along with the FAQ.