Random seed in tf.keras.models

When we build a model using tf.keras, I want the output to have a consistent value. That is to say, I do not want the output to randomly change. For example, if I run the following code, the output is always different, and I guess this may be due to the random initialization of parameters.

inputs1 = tf.keras.layers.Input(shape=(3, 1)) 
lstm1 = tf.keras.layers.LSTM(1)(inputs1)
model = tf.keras.models.Model(inputs = inputs1, outputs = lstm1)

data  = np.array([0.1, 0.2, 0.3]).reshape(1, 3, 1)
print (model.predict(data))

To resolve this problem, I observe that adding the following line at the beginning of the file solves the problem.

tf.random.set_random_seed(0)

I have two questions regarding this.

1. Is the above approach a right approach when I build a model using tf.keras?

2. If I have two models in a file, and if I want to set the random seed only for one of them, then how could I do this?

Thank you!