Size doesn't matter, success does!
Working on projects with valuable mentors that can accelerate your judgement and experience and that are successful at pleasing customers is the most important thing you can focus on career wise.
Working on project after project that is a failure will teach you things but in a backwards way that takes a very long time to understand. Working on long drawn out projects that are failing is the same thing.
Size of the project doesn't mean complex or simple. It doesn't mean more or less new things to learn. Complex and "hard" projects are usually due to mismanagement, overly complex design and architecture and the like, this can apply to multi-week projects as well as multi-year projects.
Good Enough and YAGNI
There are a great number of projects today that don't have long shelf lives. There are web projects that may only be in production for a few weeks or months depending on the lifecycle of the need of the software. Think promotional micro-sites that are temporal in nature will be thrown away and never updated or maintained.
Learning how to scale down your Architecture Astronaut tendencies and do just enough to scratch an itch is a very rare and valuable skill today.
Not that Good Enough and YAGNI don't apply to longer lifecycle projects, but scaling down process and methodology is much harder than scaling up.
Specialization is for insects
A human being should be able to change a diaper, plan an invasion,
butcher a hog, conn a ship, design a building, write a sonnet, balance
accounts, build a wall, set a bone, comfort the dying, take orders,
give orders, cooperate, act alone, solve equations, analyze a new
problem, pitch manure, program a computer, cook a tasty meal, fight
efficiently, die gallantly. Specialization is for insects.
-Robert A. Heinlein
Apply the above to software development and you won't go wrong. Being competent at many things will insulate you from fads and fetishes in the industry and ensure you are more employable than people who are one trick ponies.