Card Design
This article goes into the nuts and bolts of card design, but much of this also applies to non-card actions.
One thing
Cards and actions should do one thing.
In general, avoid making a card, action, die face, space, etc. that does two things. There are many exceptions, of course.
Instead, just make two things. That way, they can each combo with other things, and it's simpler. Cardboard costs nothing, and complexity costs a lot.
Wacky stuff
Are you designing for your own entertainment, or for the players' entertainment?
Going outside the normal rules, components, or feel of the game, or referring to things that other cards don't care about, doesn't make the game any better.
In Radlands, each player adds up the numbers on the camp cards they've chosen, and draws an opening hand of that many cards. Many people suggested or asked about a camp that draws -1 cards. That's interesting, but would it actually be fun? No. And what if they take it with two camps with 0 cards on them? Do they draw -1 cards? That would require an FAQ item.
Thinking outside the box is easy. Thinking inside the box is hard.
Conditional effects
These are effects that only work if a condition is met.
There's a huge amount of design space and depth in conditional effects. Just pair any condition with any effect. They're relatively elegant, and tie to other actions in a little chain, as the player tries to create the condition.
The Radlands expansion camp "Battering Ram" only works if you have six people.
The existence of a certain object is the easiest condition.
In Wingspan, the Eurasian Curlew lays two eggs at the end of each round, if you have a predator bird in the same habitat.
When using conditional effects, the condition must actually be something that the player can control. If it's something the game controls, then it's essentially a complex way of doing randomness.
The Radlands expansion card Death Ray used to require that you pay six water to fire it. This was a huge cost, for a huge effect. You could get six water at once, by discarding several cards with the water icon in their corner. However, you didn't actually have any control over what cards you drew. Only a few had the water icon. That was an uncontrollable condition. Later, I changed the cost to just discarding any five cards.
Even weirder are conditions that the opponent controls. The opponent will not just blithely meet the condition for you. This often just stops the opponent from doing the thing.
Several playtesters and I all invented the same camp for the Radlands expansion. It was a "super-hospital" that would restore (heal) exactly two of your injured people. In practice, however, the opponent just never left two of your people injured. That outcome was very different from the intended effect.
Self-synergies
If a player doesn't have enough control over a condition, you can give the card a second effect that helps create the condition.
The Radlands card Zeto Khan lets your event cards happen instantaneously, rather than being delayed. This is fun, but what if you don't draw any event cards? I gave him a second ability, that lets you draw three cards then discard three. This gives you a very high chance of drawing event cards.
In the base game of Radlands, you should only really choose camp cards that use punks, with ones that make punks. (Punks are random weak guys you get.) In the expansion, I wanted to add nuance to this. So, any camps that used punks, also got an extra ability that makes punks. The ability is expensive, however. This means that these camps work decently on their own, but get better if you use other punk-related camps. Choosing your camps is now far more interesting.
Rule changers
I consider the best category of designs to be the rule-changers. These cards simply make something, or the game, work differently. These can't be easily found, like the other types can. They tend to be simpler, more unique, and more thematic. Try to make more.
The new Obelisk camp from Radlands simply says "When the last card is drawn from the deck, you win."
Go through all the rules of the game, and make cards that break them.
Downsides
A good way to add a lot of strategy to cards is to create a downside, or undesirable consequence. Make cards that also affect (negatively) your own stuff. Negative effects can also be an opportunity for cleaning up the game state. The player can discard useless cards or resources.
Other sources of design
If you're operating in a very small design space (as you probably should be), it can be very hard to come up with new cards/actions. One solution I've successfully used in desperation is to steal something from a less-important part of the game.
Also, look for ways to bifurcate your existing effects into two new actions.
Give points
Many cards have an ongoing effect, and are thus useful at the start of the game, but become useless later on, as you'll have less time to use them. If your game has a "points" system, give cards with an ongoing effect a high point value. This means players will want to play those cards at the start of the game for their effect, but also at the end of the game, for their points.