I guess I just like foreshadowing in my stories. I like stories that are unresolved but them pick up again later. TV shows with fairly tight story arcs that span multiple years are pretty much my favorite stories. Babylon 5 being my favorite show ever.
To me, playing this sort of adventure has the same kind of feeling for me. It feels like someone put the time and effort in to craft a story with a definite beginning, middle, and end with threads that run through the whole thing.
In my mind, there can be loose ends left lying around, but the game absolutely needs a satisfying resolution so that the players can see, keeping with the TV example, the end credits begin to roll. In a mystery TV show, the protagonist defeats the bad guy of the week, or whatever, and maybe a big bad guy remains at large. But, they still beat that enemy. So, they can move on.
I do this all the time in my games. I've got several plot threads running. For example, in one game the PCs were attacked by assassins. Which of their enemies hired assassins to kill them? Well, they were in the middle of something else, so they decided to ignore that for then. They could have investigated it, but they prioritized other things. Eventually, a session or two later, they decided to find out what was going on with that and took care of it, and it turned out to be part of a much larger whole. However, in the meantime, they completed a bunch of other tasks.
I think what we're not seeing here is that credit roll. What has been described in this thread feels more like a cliffhanger without a resolution than a dangling plot thread to look into later. It's as if we see Captain Jean Luc Picard turned into a borg, and after the commercial break Riker decides to take a vacation on Risa and come back to Picard later after he's had another adventure. STOP THAT RIKER!!! BAD NUMBER ONE!
Why is that bad? Because that's what the "episode" was about, and you want to have a resolution at the end of the episode (or episode arc).
Now, when the episode is about something else, then it's fine to have those long plot threads, and they're very common
because they are not the main focus of each individual episode or they are part of an ongoing arc which is just equivalent to a really long episode. But, if its an arc, then you don't go off and leave that plot thread, you investigate it every single episode. Take X-Files. Every episode the viewer had their resolution. There may be signs that there's something going on in the bakground - the Smoking Man, Deep Throat, aliens, conspiracies everywhere! - but the monster of the week is taken care of. The audience needs that monster of the week to be taken care of. The foreshadowing, the conspiracy, that's background stuff. It happened, and it was interesting for the moment that it happened, but something else immediately took your attention away from it and said
"I'm important! Investigate me now or everything is going to go wrong!" and so you know that that thing that just happened isn't what you should be looking into. Because the other thing demands
immediate attention.
So, given the adventure you were running, I'd probably frame it completely differently. The PCs would resolve their adventure, catch the bad guard or whatever, and credits would roll. We'd start the next session in medias res with the guard having killed himself between sessions and the PCs talking with somebody about it, who is saying how odd it was, then immediately goblins would attack! The PCs would rush out to defend the town, but the main force would escape, probably with some kids or something. The PCs would then need to track them down right then and there, maybe rescue the kids but then discover something sinister the goblins were up to.
Given that scenario, you run into none of the problems you faced. The PCs get a resolution to the first session. The get the foreshadowing, but it is obvious that more important things are going on so they feel free to move on. They have to react immediately so there's no question of DM intent. They know that it's okay to leave the guard investigation. You're giving them implicit permission, so they can feel good about their decision to move on.
Now moving onto something a bit earlier, more the meat of what you were talking about in your OP, I believe.
I cannot find the post, but you said there was an issue where you were having an issue of expectations with the player in that he wanted to use rolls to determine success without, and I'm paraphrasing using my own words here, describing his interactions with the world and relying purely on mechanics to describe his character's actions. Would you classify this as an accurate representation of the given player's behavior?
Because, I know the topic has veered from this quite considerably, but I just wanted to point out that it might be worth it to run a one shot of Dungeon World with the player in question. I say this for a few reasons, but one of the biggest is that when changing systems, people are far more open to embracing a different playstyle. If someone has been playing D&D for a few years, and you say "Now I want you to stop playing that way and start playing like this," then there is a lot of inertia that has to be gotten over, not to mention the idea that any change means that there must be something "wrong" with the way you were playing before. But, when you're playing a brand new system, its a lot easier to say "Lets try a little more roleplay in our descriptions guys." Especially when that system is built on the idea of describing actions the way Dungeon World is.
So, if you aren't familiar, in DW players describe what they are doing in the in-game world, and they never say what roll they want to make. Never. That is 100% the purview of the GM. To attack a monster, the player has to describe how they are attacking the monster. To search a room, the player has to describe how they are searching a room. Then, the DM interprets their description as a game mechanic.
I think this would be good for the player in question to experience, and it might give him a deeper appreciation for that style of play.