All models are wrong means that no model is ever perfect. Models are used to try to explain and understand the world, but no model can perfectly explain or predict everything. For example, a model of the weather might be able to help predict if it will rain tomorrow, but it might not be able to predict if you will get a cold.