The strength of Fetch and Dual is that they are most flexible. The strength of Workshop is it taps for the most mana (and is still more flexible than Eldrazi lands). The strength of Bazaar is you need no other lands.
Eldrazi lands, even in an Eldrazi deck, are clearly 2nd tier to Workshop in an artifact deck.
I’d say Cavern of Souls is the closest printing to making a new niche, but it’s limited by the lack of great complements. You usually end up back into fetches or picking from subpar lands (like Eldrazi).
Magic mana sources are simply stuck where they are. Perhaps forever since they decided to change the entire power level of the card type about 20 years ago.
Fetches and Duals are much more powerful than most seem to realize here:
They can easily fix a 4-color manabase in a non-waste environment, a 3-color one that is waste resistant, or a 2-color one that is completely waste immune. You can even sideboard in and out of those configurations so it can vary game to game.
There are lands, like Sanctuary and Dryad Arbor, that give benefits beyond just adding mana that can and usually are played with them.
The actually ability of sac shuffle effect and deck thinning is a legitimate factor in their power level when you consider synergy with other cards, like DRS, Crucible/Loam, Brainstorm, and even random less played cards like Hermit Druid.
Overcoming the power level of these 3 factors would take some ridiculously powerful lands. The kinds of lands they haven’t printed since like Urza’s block.
The format was slower, and it was better for it at that point in Magic. That is the format that actually got me into "Vintage".
As some people have brought up there are some weird problems with rule changes though that may cause the format to be substantially different, and the restricted list still had some lag effects from it being "Type I and Type II" with a combined list. So I think the restricted list would / should be dramatically different for such a format.
I think Anti-Fish, which was just UR fish with River Boas, was actually the best deck of that time, but Hulk Smash was very powerful and so was Drain into random big drops.
Just to elaborate a bit more on that format at the time from my persepctive, UR Fish could absolutely bury you in card advantage. They not only ran Standstill, but they had Curiosity (an enchant creature!), man-lands, Null Rods, Strip effects, Grim Lavamancer. Daze, Force of Will, and the Hatchling allowed the deck to tap out every turn so they just incrementally gain massive mana usage advantage over the course of the game too. You basically had to take a strict plan of kill their early creatures, AND hit your land drops to stand any chance against them.
I don't recall the combo decks being very good. Storm was somewhat of a boogie man, but it was bad against the best deck (Fish), the restricted list made them fairly inconsistent and you had the sideboard space to devote to them if you wanted to (no Dredge). The other combos did not consistently happen early in the game, and if they did you still had a suite of 8+ counterspell effects in any control list. Occasionally, someone would get the turn 1 Dragon Combo, but you need like a perfect 7 for that.
I thought Shops was also definitely a control deck back then, not much of an aggro deck, as they mostly relied on locking you out of the game. It was a very reliant on Urza's block, whereas now people can tap into about 50 different artifact sets.
If you want a Wheel with this, you are probably going to have to play this and wheel in the same deck.
Playing Wheel of Misfortune for 0 would be fun to force your opponent to discard their hand, while mana efficient for the effect, its not that broken though given Mind Twist is a card.
The format is very saturated with good 3+ drops in control at this point so I don't think it'll have a big impact, but it obviously synergies nicely with Dack Fayden and everything else that combo'ed with Notion Thief / Narset. So there is potential.
Your opponent can kill this immediately though without you being able to recoup value, unlike Narset.